105 results about "Congestion control algorithm" patented technology

A Weighted Random Early Detection (WRED) algorithm is described. The WRED algorithm provides fairness to responsive TCP and non-responsive UDP traffic at a buffer of a communications system. Because TCP traffic is responsive to congestion occurrences while UDP traffic is not, without controls, UDP will monopolize the buffer bandwidth. This invention solves the problem by applying congestion control alogrithms to TCP and UDP traffic separately. Discard thresholds for UDP traffic are dynamically modified in a manner that limits UDP traffic to a provisioned percentage of the buffer's available bandwidth, while a traditional WRED algorithm is applied to TCP traffic.

Various embodiments of a network protocol that utilizes a congestion control algorithm that distinguishes between congestion loss and damage loss are described. In response to a packet loss on a network, a delay-based detection algorithm may be performed based on RTT (Round-Trip Time) information to determine whether the network is congested. If the delay-based detection algorithm does not determine that the network is congested then a consistency-based detection algorithm may be performed based on packet loss rate information. If either the delay-based detection algorithm or the consistency-based detection algorithm determine that the network is congested then the rate of data transmission may be reduced, e.g., by reducing a congestion window size.

A source host comprising a processing resource that supports a kernel space and a user space. A socket layer is supported by the kernel space and allows a measurement application residing in the user space to instruct a protocol layer to form a plurality of UDP test packets for transmission, from the source host, to a destination node in a communications network. The plurality of test packets employ source-based routing and an alternative congestion control algorithm is implemented by the protocol layer to that supported by the communications network.

A pipelined linecard architecture for receiving, modifying, switching, buffering, queuing and dequeuing packets for transmission in a communications network. The linecard has two paths: the receive path, which carries packets into the switch device from the network, and the transmit path, which carries packets from the switch to the network. In the receive path, received packets are processed and switched in an asynchronous, multi-stage pipeline utilizing programmable data structures for fast table lookup and linked list traversal. The pipelined switch operates on several packets in parallel while determining each packet's routing destination. Once that determination is made, each packet is modified to contain new routing information as well as additional header data to help speed it through the switch. Each packet is then buffered and enqueued for transmission over the switching fabric to the linecard attached to the proper destination port. The destination linecard may be the same physical linecard as that receiving the inbound packet or a different physical linecard. The transmit path consists of a buffer / queuing circuit similar to that used in the receive path. Both enqueuing and dequeuing of packets is accomplished using CoS-based decision making apparatus and congestion avoidance and dequeue management hardware. The architecture of the present invention has the advantages of high throughput and the ability to rapidly implement new features and capabilities.

A Weighted Random Early Detection (WRED) algorithm is described. The WRED algorithm provides fairness to responsive TCP and non-responsive UDP traffic at a buffer of a communications system. Because TCP traffic is responsive to congestion occurrences while UDP traffic is not, without controls, UDP will monopolize the buffer bandwidth. This invention solves the problem by applying congestion control alogrithms to TCP and UDP traffic separately. Discard thresholds for UDP traffic are dynamically modified in a manner that limits UDP traffic to a provisioned percentage of the buffer's available bandwidth, while a traditional WRED algorithm is applied to TCP traffic.

The present invention discloses a TCP communication protocol-based congestion window control algorithm and system, wherein the method comprises the steps of monitoring the actual throughput and the continuous packet loss number of a network; according to the actual throughput of the network and a desired throughput, obtaining the network time delay change; determining whether the network time delay change is greater than a first threshold value and whether the continuous packet loss number is greater than a second threshold value; if yes, determining that the network congests, and updating a slow start threshold value according to a measured network available bandwidth; and setting the size of a congestion window according to the slow start threshold value. The method can determine whether the network congests accurately according to the network time delay change and the packet loss property, and can set the congestion window according to a real-timely monitored bandwidth estimation value when the network is determined to congest, so that the congestion window can converge to an actual bandwidth faster, the utilization rate of the bandwidth resources is improved, and the user experience is improved based on the operation speed of a TCP application.

Methods and apparatus are provided to improve data throughput in a wireless, wireline or a combination wireless and wireline communication system. A congestion control manager selects between an assumption based congestion control algorithm and a speculation based congestion control algorithm. The selected algorithm generates data recovery instructions including instructions for resizing, or not, congestion window sizing for the communication gateways. By making the selection between the assumption based congestion control algorithm and the speculation based congestion control algorithm based upon network information, data recovery and throughput is optimized for networks having lossy data links.

The invention discloses a method for adjusting a sending rate in a bandwidth detection stage and a congestion control algorithm, and the method comprises the steps: firstly increasing the sending ratethrough a speed increasing factor g in a first RTT in the bandwidth detection stage, and then sending a round of data package, so as to carry out the bandwidth detection; judging the congestion condition of the current network according to whether ACK packets of all data packets sent in the round are received within a certain period of time; when the condition of the current network is good, adjusting the acceleration factor of the sending rate according to the current round-trip delay and the historical maximum transmission rate; when the condition of the current network is poor, reducing the sending rate to send a round of data packets, i.e., reducing the sending rate to empty the queue in the second RTT of the bandwidth detection stage; and in the remaining period of the bandwidth detection stage, determining the sending rate according to the real-time transmission rate. The method can give consideration to the transmission efficiency, convergence and fairness.

The invention discloses a delay-based adaptive congestion control method, which comprises the following steps of: judging whether a network is congested or not at a sending end according to queuing delay and a window size, and adjusting a congestion window according to a congestion condition, wherein the adjustment amplitude of the congestion window each time is V * theta * delta, delta is a constant, V is used for accelerating the convergence rate of the congestion control algorithm, and theta is a parameter adjusted according to the situation whether zero point is reached or not in the queuing delay periodic jitter process, wherein the parameter is used for controlling the window adjusting amplitude in the current network environment, so that the problems that the link utilization rate is reduced under the condition that the long-term queuing delay is 0 and the minimum round-trip delay cannot be measured when the queuing delay is not 0 for a long time are solved. Experimental resultsshow that the bottleneck link buffer queue can be controlled to periodically generate emptying behaviors, so that each stream can measure accurate round-trip link propagation delay. Compared with a Copa algorithm, the method has the advantages of being lower in delay and better in fairness.

The invention discloses a satellite network congestion control method based on bandwidth estimation, comprising the following steps: S1: establishing a network transmission link; S2: using the Hybla-BE congestion control algorithm for data communication; S3: the TCP sending end of the ground station 1 Calculate the estimated bandwidth according to the receiving frequency of the received ACK; S4: Calculate the expected bandwidth; S5: Calculate the remaining bandwidth of the communication network, the congestion window growth factor ρcurrent of the Hybla algorithm, and the congestion window growth factor ρnew of the Hybla-BE algorithm; S6: According to The adjusted window growth factor ρnew adjusts the congestion window value; S7: Adjusts the congestion window value: the TCP sending end of ground station I predicts the bandwidth environment of the network by dynamically estimating the bandwidth value, and scales and adjusts the window growth factor according to the remaining bandwidth to achieve congestion The growth of the window value can dynamically adapt to the network bandwidth condition.

A source host comprising a processing resource that supports a kernel space and a user space. A socket layer is supported by the kernel space and allows a measurement application residing in the user space to instruct a protocol layer to form a plurality of UDP test packets for transmission, from the source host, to a destination node in a communications network. The plurality of test packets employ source-based routing and an alternative congestion control algorithm is implemented by the protocol layer to that supported by the communications network.